Daily streamflow forecasting using a wavelet transform and artificial neural network hybrid models

Abstract

New wavelet and artificial neural network (WA) hybrid models are proposed for daily streamflow forecasting at 1, 3, 5 and 7 days ahead, based on the low-frequency components of the original signal (approximations). The results show that the proposed hybrid models give significantly better results than the classical artificial neural network (ANN) model for all tested situations. For short-term (1-day ahead) forecasts, information on higher-frequency signal components was essential to ensure good model performance. However, for forecasting more days ahead, lower-frequency components are needed as input to the proposed hybrid models. The WA models also proved to be effective for eliminating the lags often seen in daily streamflow forecasts obtained by classical ANN models. 

Editor D. Koutsoyiannis; Associate editor L. See

Citation Santos, C.A.G. and Silva, G.B.L., 2013. Daily streamflow forecasting using a wavelet transform and artificial neural network hybrid models. Hydrological Sciences Journal, 59 (2), 312–324.     

Zooplankton succession during extraordinary drought–flood cycles: A case study in a South American floodplain lake

We examined the zooplankton abundance and composition of Laguna Grande, a floodplain wetland of the Lower Paraná Basin (Argentina), during an extraordinary drought–flood cycle that affected both the environment and the biological conditions of the lake. Low waters were characterised by remarkably high conductivities and pH values, and high phytoplankton and bacterioplankton abundances with cyanobacterial blooms, while high waters showed opposite features. In relation to zooplankton, the mean abundances of all the taxonomic groups (rotifers, cladocerans, copepods, ciliates, and heterotrophic nanoflagellates) were slightly higher at low waters. Major changes were observed in the specific composition of metazooplankton: the euryhaline species assemblage that dominated in the dry warm period was replaced by several oligohaline littoral and planktonic species characteristic of the Paraná River Basin, when the water level rose. Mean species richness values at high waters doubled those of low waters and were directly correlated to water depth. Most of the rotifers of the genus Brachionus and the cladoceran Moina micrura switched from parthenogenetic to sexual reproduction during low waters, as a response to a harsh environment and crowding. We suggest that the main changes in the environmental conditions in this eutrophic floodplain lake are driven by the hydrology, which regulates the zooplankton succession. The herein described shifts in the zooplankton structure and dynamics of Laguna Grande over an extraordinary drought–flood cycle contribute to the understanding of the processes that might occur under the scenarios predicted by climate change models.     

Cenozoic tectonics on the Galicia margin,northwest Spain

Neogene tectonic activity (Betic Orogeny) along the southern plate boundary of Iberia, as it converged with Africa and the Alboran Plate was inserted between them, propagated along Iberia's west side as far north as Galicia, northwest Spain. As the activity propagated northward, it reactivated structures formed during the Paleozoic (Hercynian), the Early Cretaceous opening of the northern North Atlantic and the Bay of Biscay, and the Paleogene Pyrenean Orogeny when Iberia collided with Eurasia. Recent earthquakes indicate that this tectonism is still active today as far north as northwest Spain.     

Entropy inference of recurrence models for variability analysis of extreme events

A complete methodology is developed to analyze the recurrence of extreme environmental events and its variability as time without further events elapses. Firstly we investigate the conditioned recurrence inference problem consisting in the selection of a probability model for the interarrival time between extreme events, given a contexto-factual evidence conditioned by the time elapsed since the last of such events. Two ways to include this condition can be considered, which yield alternative conditioned evidences and convert the former problem into two distinct ones, thus giving rise to a possible consistency violation. These problems are formalized within the logical probability framework, in a plausible logic language that allows a suitable expression of the available observational data. They are solved using the REF relative entropy method with fractile constraints, and their solutions are compared at all inference levels. It is concluded that the two conditioning ways are not really mutually exclusive and that a unique global solution to the conditioned inference can be obtained using this procedure. An example illustrates an application of the methodology to the variability analysis of the recurrence time between historical inundations of the Guadalquivir river in Spain, as time elapses with no new floods.Acknowledgments. Support for this work was provided by DGI of Spain as the grant REN2000-2988-E/CLI and the research project REN2002-01337/CLI.     

Optimization of Time Data Codification and Transmission Schemes: Application to Gaia

Gaia is an ambitious space observatory devoted to obtain the largest and most precise astrometric catalogue of astronomical objects from our Galaxy and beyond. On-board processing and transmission of the huge amount of data generated by the instruments is one of its several technological challenges. The measurement time tags are critical for the scientific results of the mission, so they must be measured and transmitted with the highest precision – leading to an important telemetry channel occupation. In this paper we present the optimization of time data, which has resulted in a useful software tool. We also present how time data is adapted to the packet telemetry standard. The several communication layers are illustrated and a method for coding and transmitting the relevant data is described as well. Although our work is focused on Gaia, the timing scheme and the corresponding tools can be applied to any other instrument or mission with similar operational principles.     

Uranium-rich accessory minerals in the peraluminous and perphosphorous Belvís de Monroy pluton (Iberian Variscan belt)

The strongly peraluminous, perphosphorous (<0.85 wt% P₂O₅) and low-Ca granites from the Belvís de Monroy pluton contain the most U-rich monazite-(Ce) and xenotime known in igneous rocks. Along with these accessory minerals, P-rich zircon occurs, reaching uncommon compositions particularly in the more fractionated units of this zoned pluton. Monazite displays a wide compositional variation of UO₂ (<23.13 wt%) and ThO₂ (<19.58 wt%), positively correlated with Ca, Si, P, Y and REE. Xenotime shows a high UO₂ content (2.37–13.34 wt%) with parallel increases of LREE, Ca and Si. Zircon contains comparatively much lower UO₂ (<1.53 wt%) but high P₂O₅ (<14.91 wt%), Al₂O₃ (<6.96 wt%), FeO (<2.93 wt%) and CaO (<2.24 wt%) contents. The main mechanism of incorporating large U and Th amounts in studied monazite and U in xenotime is the cheralite-type [(Th,U)⁴⁺ + Ca²⁺ = 2(Y,REE)³⁺] substitution. Zircon requires several coupled mechanisms to charge balance the P substitution, resulting in non-stoichiometric compositions with low analytical totals. Compositional variations in the studied accessory phases indicate that the substitution mechanisms during crystal growth depend on the availability of non-formula elements. The strong P-rich character of the studied granites increases monazite crystallization, triggering a progressive impoverishment in Th and LREE in the residual melts, and consequently increasing extraordinarily the U content in monazite and xenotime. This is in marked contrast to other peraluminous (I-type or P-poor S-type) granite series. The P-rich and low-Ca peraluminous melt inhibits uraninite crystallization, so contributing to the U availability for monazite and xenotime.     

Spatiotemporal variability of vegetation due to drought dynamics (2012–2017): a case study of the Upper Paraíba River basin,Brazil

Natural Hazards - Vegetation indices have been widely used for monitoring the spatiotemporal variables of vegetation and characterizing droughts, primarily in semiarid regions. Drought is a...     

“Global warming” versus “climate change”: A replication on the association between political self-identification,question wording,and environmental beliefs

Climate change is one of the greatest threats to humanity and requires immediate action. Schuldt, Konrath, and Schwarz (2011) suggested that beliefs in environmental phenomena can be influenced by the terminology used to describe it: changing question wording from global warming to climate change resulted in a 6.3 percentage point increase in belief in environmental phenomena. This association was moderated by political self-identification, with Republicans being 16.2 percentage points more likely to believe in climate change than in global warming, with Democrats showing no difference. The potential for connotative meanings to shift over time and the sociopolitical changes since the original study, potential policy and environmental campaign implications, and an expansion of these findings to other countries, motivated an attempt to replicate this important finding. This pre-registered study repeated the original procedures in the United States of America and two other countries (United Kingdom and Australia; total N = 5,717). Although question wording no longer had a significant effect on beliefs in climate change/global warming, the association of political self-identification with beliefs in environmental phenomena replicated in all three countries, with Conservatives consistently believing less in climate change/global warming than Liberals. The potential impacts of temporal and methodological differences on the discrepancies between this study's and the original's findings are discussed.     

The boron isotope geochemistry of tourmaline-rich alteration in the IOCG systems of northern Chile: implications for a magmatic-hydrothermal origin

Hydrothermal tourmaline is common in the iron oxide-copper-gold (IOCG) deposits of the Coastal Cordillera of Chile where it occurs as large crystals in the groundmass of magmatic-hydrothermal breccias, such as in the Silvita or Tropezón ore bodies, or as small grains in replacive bodies or breccia cement in the ore-bearing andesite, as seen at the Candelaria or Carola deposits. Tourmaline shows strong chemical zoning and has a composition of schorl–dravite with significant povondraite and uvite components. The observed boron isotope composition is fairly variable, between −10.4‰ and +6.0‰ with no major differences among the different deposits, suggesting a common genetic mechanism. The δ¹¹B values are significantly lower than those of seawater or marine evaporites and very similar to those of younger porphyry copper deposits and volcanic rocks in the region, indicating that the boron has a common, likely magmatic, origin. The predominant boron source was ultimately dewatering of the subducting slab with a significant contribution derived from the overlying continental basement. The range of δ¹¹B values is between those of the porphyry copper deposits and the porphyry tin deposits of the Andes, suggesting that the IOCG mineralization might be genetically related to fluids having more crustal contamination than the porphyry copper deposits; such an interpretation is at odds with current models that propose that the Andean IOCG deposits are related to juvenile melts or to the circulation of basinal brines. Furthermore, the obtained δ¹¹B data are markedly different from those of the tourmaline in the Carajás IOCG district (Brazil), suggesting that IOCGs do not form by a unique mechanism involving only one type of fluids.